MADS78 and MADS79 Are Essential Regulators of Early ...MADS78 and MADS79 Are Essential Regulators of...

MADS78 and MADS79 Are Essential Regulators of EarlySeed Development in Rice1[OPEN]

Puneet Paul,a Balpreet K. Dhatt ,a Michael Miller,a,2 Jing J. Folsom,a Zhen Wang,a,3 Inga Krassovskaya,b

Kan Liu,c Jaspreet Sandhu,a Huihui Yu,c Chi Zhang,c Toshihiro Obata,b Paul Staswick,a andHarkamal Waliaa,4,5

aDepartment of Agronomy and Horticulture, University of Nebraska, Lincoln, Nebraska 68583bDepartment of Biochemistry, University of Nebraska, Lincoln, Nebraska 68588cSchool of Biological Science, University of Nebraska, Lincoln, Nebraska 68588

ORCID IDs: 0000-0001-8220-8021 (P.P.); 0000-0002-5420-7387 (J.J.F.); 0000-0003-1015-4460 (I.K.); 0000-0003-2725-1937 (H.Y.);0000-0002-1827-8137 (C.Z.); 0000-0001-8931-7722 (T.O.); 0000-0003-2798-0275 (P.S.); 0000-0002-9712-5824 (H.W.).

MADS box transcription factors (TFs) are subdivided into type I and II based on phylogenetic analysis. The type II TFsregulate floral organ identity and flowering time, but type I TFs are relatively less characterized. Here, we report thefunctional characterization of two type I MADS box TFs in rice (Oryza sativa), MADS78 and MADS79. Transcriptabundance of both these genes in developing seed peaked at 48 h after fertilization and was suppressed by 96 h afterfertilization, corresponding to syncytial and cellularized stages of endosperm development, respectively. Seedsoverexpressing MADS78 and MADS79 exhibited delayed endosperm cellularization, while CRISPR-Cas9-mediatedsingle knockout mutants showed precocious endosperm cellularization. MADS78 and MADS79 were indispensablefor seed development, as a double knockout mutant failed to make viable seeds. Both MADS78 and 79 interactedwith MADS89, another type I MADS box, which enhances nuclear localization. The expression analysis of Fie1,a rice FERTILIZATION-INDEPENDENT SEED-POLYCOMB REPRESSOR COMPLEX2 component, in MADS78 and79 mutants and vice versa established an antithetical relation, suggesting that Fie1 could be involved in negativeregulation of MADS78 and MADS79. Misregulation of MADS78 and MADS79 perturbed auxin homeostasis andcarbon metabolism, as evident by misregulation of genes involved in auxin transport and signaling as well asstarch biosynthesis genes causing structural abnormalities in starch granules at maturity. Collectively, we show thatMADS78 and MADS79 are essential regulators of early seed developmental transition and impact both seed sizeand quality in rice.

Double fertilization is a characteristic feature of an-giosperm reproduction. During double fertilization,two haploid sperm cells reach the embryo sac via thepollen tube. One sperm cell fuses with a haploid egg cellto form a diploid embryo and the other fuses with adiploid central cell to form a triploid endosperm cellcomposed of two maternal genomes and one paternalgenome (Olsen, 2004). The embryo acts as a precursor tothe next generation of the plant life cycle, and the en-dosperm is the terminal nutritive tissue. In eudicots, theembryo represents the majority of the tissue in matureseeds and the endosperm is ephemeral (Chaudhuryet al., 2001). However, in monocots such as rice(Oryza sativa), wheat (Triticum aestivum), and maize(Zea mays), the endosperm persists and becomes thedominant tissue that stores starch and proteins, thusdetermining the mature seed size (Chaudhury et al.,2001; Gao et al., 2013). Endosperm is the primarysource of calories for human (Homo sapiens) and live-stock nutrition, thereby making it a valuable biologicaland economic entity (Sabelli and Larkins, 2009).Endosperm development undergoes several distinct

and overlapping phases categorized as follows: (1)early, the triploid nuclei undergo multiple rounds of

1This work was supported by the National Science Foundation(grant no. 1736192 to H.W.).

2Present address: School of Integrative Plant Science, Cornell Uni-versity, Ithaca, NY 14853.

3Present address: Institute of Animal Science, Chinese Academy ofAgricultural Sciences, 100193 Beijing, China.

4Author for contact: [email protected] author.The author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:Harkamal Walia ([email protected]).

P.P. and H.W. conceived and designed the study; P.P. generatedsingle and double knockouts of MADS78 and 79; J.J.F. generatedoverexpression mutants of MADS78 and 79 and Fie1; B.K.D. andJ.S. generated single knockout of Fie1; P.P. performed experimentson MADS78 and 79 mutants; B.K.D. analyzed Fie1 mutants and per-formed RT-qPCR; Z.W. performed in situ hybridization assay; inter-action assays were performed by M.M. and P.P.; T.O. and I.K.performed metabolite profiling; K.L. and C.Z. analyzed RNA-seq;H.Y. performed target gene analysis; P.S. critically reviewed and an-alyzed the work; P.P. analyzed the results from all the experimentsand wrote the article; all authors read and approved the article.

[OPEN]Articles can be viewed without a subscription.www.plantphysiol.org/cgi/doi/10.1104/pp.19.00917

Plant Physiology�, February 2020, Vol. 182, pp. 933–948, www.plantphysiol.org � 2020 American Society of Plant Biologists. All Rights Reserved. 933

https://plantphysiol.orgDownloaded on March 14, 2021. - Published by Copyright (c) 2020 American Society of Plant Biologists. All rights reserved.

https://orcid.org/0000-0001-8220-8021

https://orcid.org/0000-0001-8220-8021

https://orcid.org/0000-0002-5420-7387

https://orcid.org/0000-0002-5420-7387

https://orcid.org/0000-0003-1015-4460

https://orcid.org/0000-0003-1015-4460

https://orcid.org/0000-0003-2725-1937

https://orcid.org/0000-0003-2725-1937

https://orcid.org/0000-0002-1827-8137

https://orcid.org/0000-0002-1827-8137

https://orcid.org/0000-0001-8931-7722

https://orcid.org/0000-0001-8931-7722

https://orcid.org/0000-0003-2798-0275

https://orcid.org/0000-0003-2798-0275

https://orcid.org/0000-0002-9712-5824

https://orcid.org/0000-0002-9712-5824

https://orcid.org/0000-0001-8220-8021

https://orcid.org/0000-0002-5420-7387

https://orcid.org/0000-0003-1015-4460

https://orcid.org/0000-0003-2725-1937

https://orcid.org/0000-0002-1827-8137

https://orcid.org/0000-0001-8931-7722

https://orcid.org/0000-0003-2798-0275

https://orcid.org/0000-0002-9712-5824

http://crossmark.crossref.org/dialog/?doi=10.1104/pp.19.00917&domain=pdf&date_stamp=2020-01-21

http://dx.doi.org/10.13039/100000001

mailto:[email protected]

http://www.plantphysiol.org

mailto:[email protected]

http://www.plantphysiol.org/cgi/doi/10.1104/pp.19.00917

https://plantphysiol.org

mitotic divisions without cytokinesis (syncytium) be-fore initiating cell wall formation (cellularization); (2)mid, differentiation of cells, grain filling that is pri-marily contributed by synthesis of starch and proteins,and endoreduplication; and (3) late, maturation anddehydration (Sabelli and Larkins, 2009). The rate andduration of endosperm proliferation during the syn-cytial phase is an important determinant of final seedsize (Garcia et al., 2005). Normal endosperm develop-ment requires a timely transition from syncytiumto cellularization (Brown et al., 1996; Olsen, 2004). Ge-nomic imbalance between the parental genomes dis-rupts the timely initiation of cellularization (Brownet al., 1996; Haig, 2013; Wang et al., 2018). Excessivepaternal dosage delays cellularization and causes en-dosperm overproliferation, resulting in bigger seeds(Scott et al., 1998; Stoute et al., 2012). Conversely, in-creased maternal dosage leads to precocious cellulari-zation and smaller seeds. Either case may result in seedabortion as well (Scott et al., 1998; Stoute et al., 2012).

The transition of endosperm from the syncytial tocellularized state is regulated at both the genetic andepigenetic levels and is sensitive to environmentalconditions (Folsom et al., 2014). Epigenetic regulation ismediated either byDNAmethylation or FERTILIZATION-INDEPENDENT SEED-POLYCOMB REPRESSIVECOMPLEX2 (FIS-PRC2; Gehring, 2013; Pires, 2014).METHYLTRANSFERASE1 (MET1) predominantlymaintains CG DNA methylation, and a cross betweenmet1 as pollen donor and the wild type activates oth-erwise repressed maternal-specific genes (Finneganet al., 1996; Genger et al., 1999; Köhler et al., 2012).The FIS-PRC2 complex regulates endosperm develop-ment by installing a repressive chromatin mark, tri-methylation of histone H3 at Lys-27 (H3K27me3), onseveral MADS box and other seed development-relatedgenes (Huh et al., 2007; Folsom et al., 2014; Zhang et al.,2018). MADS box transcription factors (TFs), dividedinto type I and type II (or MIKC type), are known fortheir role in floral organ identity and flowering timecontrol. The type II genes have been extensively studiedas floral homeotic genes, while only a few type I geneshave been functionally characterized (Bemer et al.,2010; Masiero et al., 2011; Smaczniak et al., 2012;Chen et al., 2016). In Arabidopsis (Arabidopsis thaliana),a well-studied type I MADS box TF, AGAMOUSLIKE62 (AGL62) shows specific syncytial phase ex-pression that declines with the progression of cellu-larization; the agl62 mutant undergoes precociousendosperm cellularization (Kang et al., 2008). Severaltype I MADS box genes in Arabidopsis are impli-cated in the fis-prc2 mutants, which fail to undergoendosperm cellularization (Chaudhury et al., 1997;Grossniklaus et al., 1998; Köhler et al., 2003;Walia et al.,2009; Zhang et al., 2018). Interestingly, genes encod-ing the components of the FIS-PRC2 complex are alsoimprinted and prone to misregulation due to paternalexcess (Luo et al., 2000, 2009; Jullien and Berger, 2010).Thus, normal seed development requires a multifac-eted interplay between parental genome dosage, the

imprinting of FIS-PRC2 components, and the role ofFIS-PRC2 in the imprinting of downstream factors(Pires, 2014).

Here, we report the functional characterization oftwo rice type I MADS box genes, MADS78 andMADS79, that are active during early seed develop-ment. We report that MADS78 and MADS79 are es-sential for normal seed development. Misregulation ofeither gene results in seed abnormalities due to mis-timed developmental transition of the endosperm anddownstream impacts on starch/grain quality. At leastone of the two MADS genes is required for rice seedviability, suggesting partial functional redundancy.Overall, our work shows that MADS78 and MADS79regulate agronomically important yield traits, includ-ing grain size, fertility, and grain quality, in rice.

RESULTS

Expression Analysis of MADS78 and 79

Based on phylogenetic analysis, type I MADS boxgenes are divided into three classes: Ma, Mb, and Mg(Parenicová et al., 2003). The rice genome has 32 type IMADS box proteins composed of 13 Ma, nine Mb, and10 Mg proteins (Arora et al., 2007). Both MADS78 andMADS79 correspond to the Ma subclade and exhibithigh sequence similarity (DNA, 83% and protein, 69%;Supplemental Fig. S1). The expression profile of thesetwo genes in public databases was limited, as they werenot represented on the microarray platforms. The RNAsequencing (RNA-seq)-based expression analysis alsopresents a challenge due to high sequence similarity, asone member is biased for mapping of most reads(Supplemental Table S1). Therefore, we determined thespatial and temporal expression patterns of MADS78and MADS79 using reverse transcription quantitativePCR (RT-qPCR) assays and in situ hybridizations(Fig. 1). Both genes were preferentially expressed inpollen and ovary (Fig. 1A). Furthermore, tissue-specificexpression patterns determined by in situ hybridizationassays indicated that both genes were expressed inpollen grains as well as in anther walls (Fig. 1C). Indeveloping seeds, the transcript abundance of bothgenes peaked at 48 h after fertilization (HAF), coincid-ing with the syncytium stage, where endosperm un-dergoes rapid nuclear division without forming cellwalls (Fig. 1B). The transcript levels of both genes de-clined by 72 HAF and were significantly reduced at 96HAF, which coincidedwith the onset and completion ofendosperm cellularization, respectively (Fig. 1B). Thisdecline in transcript levels was also confirmed viain situ hybridization assays where sense/antisenseprobes for both genes were used during each devel-opmental stage (Fig. 1C). Furthermore, the preferentialexpression pattern of both genes during endospermdevelopment was supported by the presence of regu-latory motifs in their promoters (Supplemental Fig. S2).RY-element, Skn-1, and GCN4 motifs are known to

934 Plant Physiol. Vol. 182, 2020

Paul et al.


http://www.plantphysiol.org/cgi/content/full/pp.19.00917/DC1




establish seed- or endosperm-specific expression(Dickinson et al., 1988; Blackwell et al., 1994; Yoshiharaet al., 1996).

Both MADS78 and 79 Heterodimerize with MADS87and MADS89

Type I MADS box proteins in Arabidopsis havea well-characterized interactome exhibiting exten-sive heterodimerization (de Folter et al., 2005). Wenext tested the nature and extent of interactions ofMADS78 and MADS79 in rice. For this, we clonedthe coding sequences (CDS) of 13 rice type I MADSbox proteins from all three subclades: four Ma(MADS70, MADS77, MADS78, and MADS79), threeMb (MADS90, MADS96, and MADS98), and six Mg(MADS81, MADS83, MADS84, MADS87, MADS88, andMADS89). The resulting prey clones were cotrans-formed with MADS78 or MADS79 as bait in a bimolec-ular fluorescence complementation (BiFC) interactionassay. For controls, separate cotransformation ofMADS78 and 79 was performed with an unrelatedprotein, bZIP76, as well as with an empty vector(Supplemental Fig. S3). Two Mg member proteins,MADS87 and MADS89, showed positive interactionswith both MADS78 and MADS79 (Fig. 2). Positive in-teractions were also observed when we cotransformedMADS87 or MADS89 with MADS78 or MADS79 byswapping the respective vectors to test the veracity ofthese interactions (Supplemental Fig. S4). NeitherMADS78 nor MADS79 formed homodimers.Since both MADS78 and MADS79 are TFs, they are

expected to localize to the nucleus. To test this possi-bility, we fused GFP to the N or C terminus of the CDSfor both genes. TheN-terminal GFP fusion constructs of

both genes localized to the nucleus as well as to thecytosol (Fig. 3). The C-terminal GFP fusion constructsfor MADS78 localized exclusively to the nucleus, whileMADS79 localized around chloroplasts (SupplementalFig. S5A). In silico analysis revealed that neither genecarries a nuclear localization signal (NLS), suggestingthat both genes likely require a recruiting factor oran anchor for nuclear transportation (SupplementalTable S2). Since both genes interact with the sameMg proteins (MADS87 and MADS89; Fig. 2), we co-transformed the nonfluorescent BiFC construct (YFN43)encoding a positive interaction partner (MADS87 orMADS89) with MADS78 and MADS79 GFP fusions.WhenMADS78 (fusedwith GFP on its N terminus) wascotransformedwithMADS89, the GFP-MADS78 fusionprotein specifically localized to the nucleus (Fig. 3).Similarly, MADS79 fused with GFP on its N or C ter-minus and cotransformed with MADS89 specificallylocalized to the nucleus (Fig. 3; Supplemental Fig. S5A).However, cotransformation of MADS78 or MADS79with the other positive interactor, MADS87, failed toalter their localization (Supplemental Fig. S5, B–D).

Overexpression of MADS78 and 79 Leads to High SpikeletSterility and Delayed Cellularization

To functionally characterize MADS78 and MADS79,we generated ubiquitin-promoter-driven overexpression(OE) plant lines (OE78-4/OE78-8 and OE79-3/OE79-4for MADS78 and MADS79, respectively; Fig. 4). Wedid not detect any phenotypic differences between OEand wild-type plants during the vegetative phase(Supplemental Fig. S6). However, we observed signif-icant spikelet sterility in most of the OE lines for bothgenes compared with the wild type (Fig. 4, B and C).

Figure 1. Expression analysis ofMADS78andMADS79. A and B, RT-qPCR analysisof MADS78 and MADS79 in maturepollen and unfertilized ovary (A) andearly seed development (48, 72, and96 HAF; B) relative to leaf tissue fromwild-type plants. For statistical analysis,Student’s t test was used, and differentlowercase letters indicate significantdifferences (P , 0.05). Error bars indi-cate SD (n5 6; two biological and threetechnical replicates). C, In situ hybridi-zation of MADS78 and MADS79 inmature anthers, unfertilized ovary, andseeds at 48 and 72 HAF. aw, Antherwall; cv, central vacuole; pg, pollengrain. Bars 5 0.2 mm.

Plant Physiol. Vol. 182, 2020 935

The Role of MADS78 and MADS79 in Rice Seed Development













Seed sterility was more severe for OE78 compared withOE79 plants, where sterility was significantly higheronly for one plant line (OE79-4) relative to the wild type(Fig. 4C). To determine the basis of seed sterility, weassayed the viability of mature pollen derived fromOE plants. We did not find significant differences inpollen viability between OE and wild-type plant lines(Supplemental Fig. S7). Furthermore, careful inspec-tion of the dehusked sterile seeds at maturity revealedflattened seed mass instead of aborted ovaries, sug-gesting postzygotic failure as the cause of the seed

sterility phenotype (Supplemental Fig. S8). This ob-servation prompted us to investigate postzygotic seeddevelopment during early stages (48, 72, and 96 HAF)when both MADS78 and MADS79 were preferen-tially expressed (Fig. 1B). We observed that developingseeds from OE lines could be categorized into twotypes: abnormal and normal. The abnormally devel-oping seeds from the OE lines were distinguished bytheir smaller size relative to the normal-looking seeds,which were indistinguishable from the developingseeds of wild-type plants (Supplemental Fig. S9A). Thepercentage of normally developing seeds in the OElines was significantly lower than in the wild type(Supplemental Fig. S9B). Consistent with the spikeletsterility phenotype, the lower percentage of normaldeveloping seeds in the OE78 lines was much moresevere than in the OE79 lines (Fig. 4C; SupplementalFig. S9B). This suggested that the severity of sterility

Figure 2. Both MADS78 and MADS79 heterodimerize withMADS87 and MADS89. BiFC assay shows positive interaction ofMADS78 (A) and MADS79 (B) with MADS87 and MADS89. For acontrol, we cotransformed both constructs with an unrelated gene(bZIP76). Bars 5 25 mm.

Figure 3. MADS89 enhances the specificity of nuclear localization forMADS78 and MADS79. In the absence of MADS89, N-terminal GFPfusion constructs of MADS78 and MADS79 localize to nucleus andcytosol. Cotransformation of the respective constructs with MADS89enhances their localization specifically to the nucleus. Empty vectorwas used as a control. Bar 5 25 mm.


Paul et al.









observed at maturity is determined during early seeddevelopment when both MADS78 and MADS79 aretranscriptionally active. We also observed that fullydeveloped mature seeds derived from the OE plantlines had significantly higher length-to-width ratiosrelative to wild-type seeds at maturity (Table 1). Weobserved decreases in 1,000-grain weight in both OElines; however, the decrease was only significant forOE78 lines (Table 1). To further investigate the devel-opmental defects caused by misregulation of these twogenes, we examined cross sections of developing seedsfrom OE and wild-type plants. Seeds from wild-typeplants completed endosperm cellularization by 96 HAF(Fig. 4D). However, seeds from OE MADS78 andMADS79 lines did not completely cellularize by 96HAF.This suggests that the higher abundance of MADS78 orMADS79 and/or their ubiquitous expression causesdelayed endosperm cellularization (Fig. 4D).

Single Knockout Mutants of MADS78 or MADS79 ShowPrecocious Endosperm Cellularization

For each of the genes, we selected two homozygoussingle mutants (MADS78, mads78_17 and mads78_19;

MADS79, mads79_23 and mads79_24). Both mutantlines corresponding to MADS78 had deletions in thetargeted regions and resulted in truncated transcriptsdue to premature stop codons (Supplemental Fig. S10A).MADS79 mutants, mads79_23 and mads79_24, carriedan insertion of adenine and a 4-bp deletion, respectively(Supplemental Fig. S10A). Neither of the MADS79mutations led to premature stop codons. RT-qPCRanalysis showed a decline in transcript abundance ofMADS78 and MADS79 in the mutants relative to thewild type at 48 HAF (Supplemental Fig. S10, B and C).At reproductive maturity, we did not observe severespikelet sterility in the mutants compared with the wildtype (Supplemental Fig. S10, D and E). Like the OElines, one single knockout mutant line of both genes(mads78_19 and mads79_23) showed a significant in-crease in length-to-width ratio of mature seeds relativeto the wild type (Table 1). We detected significant de-creases in 1,000-grain weight in single knockout lines ofboth MADS78 and MADS79 (Table 1). Next, we per-formed histochemical analysis of developing seedsto check the timing of endosperm cellularization.The knockout mutants exhibited a faster rate of endo-sperm cellularization relative to the wild type (Fig. 5).As a result, knockout mutants underwent complete

Figure 4. Overexpression ofMADS78 andMADS79 leads to partial spikelet sterility and delayed endosperm cellularization. A,RT-qPCR analysis ofMADS78 andMADS79 expression in 11-d-old seedlings and developing seeds (48 and 72 HAF) in the wild-type (WT) and OE mutants for the respective tissue. Wild-type expression was used as a baseline for the corresponding tissue andgene. B, Representative mature panicle images of the OE mutants compared with the wild type. White arrows indicate sterileseeds. Images shown were digitally extracted and scaled for comparison. Bar5 2 cm. C, Quantification of spikelet fertility (%) inthe wild type and OE mutants; n5 25 to 40 plants. D, Representative cross sections of developing seeds at 96 HAF from the wildtype andOEmutants. cv, Central vacuole; ed, endosperm. Bar5 0.2mm. For statistical analysis in A and C, Tukey’s test was used:***P , 0.001 and **P , 0.01. Error bars indicate SD (n 5 6; two biological and three technical replicates).









endosperm cellularization by 84HAF, whereas thewildtype completed it by 96 HAF (Fig. 5). This suggests thatseeds deficient in either of the two genes exhibit pre-cocious endosperm cellularization.

The Double Knockout Mutation for MADS78 and 79Is Lethal

Given the high percentage of sequence similarity andoverlap in MADS78 and MADS79 gene activity duringrice reproductive and seed development, we next ex-amined the combined role of MADS78 and MADS79.For this, we generated CRISPR-Cas9-mediated doubleknockout mutants. We obtained two homozygous T0mutants corresponding to each of the single guideRNAs (sgRNAs): sgRNA-1 (mads78-79_5 and mads78-79_9) and sgRNA-2 (mads78-79_3 and mads78-79_4;Supplemental Fig. S11A). Both mads78-79_5 andmads78-79_9 double mutants (sgRNA-1) had a homo-zygous point mutation within both genes (insertion ofadenine). In addition, mads78-79_9 possessed a substi-tution (guanine to thymine) in both alleles of MADS78(Supplemental Fig. S11A). The mads78-79_3 (sgRNA-2)double knockout mutants had a homozygous insertionof an adenine in both genes, while mads78-79_4 hadan insertion of thymine and guanine in MADS78and MADS79, respectively (Supplemental Fig. S11A).All double knockout mutant lines were completelysterile, as they failed to develop mature seeds (Fig. 6;Supplemental Fig. S11B). On the other hand, weobtained seed from double knockout mutants that wereheterozygous for one or both genes (SupplementalFig. S11C). These data indicated that at least one func-tional gene (MADS78 or MADS79) is required for seedviability in rice.

Expression of MADS78 and MADS79 Is NegativelyAssociated with Fie1

Several type I MADS box genes in Arabidopsis(Walia et al., 2009; Zhang et al., 2018; Bjerkan et al.,2019) and rice (Folsom et al., 2014; Chen et al., 2018;

Wang et al., 2018) are imprinted. Therefore, we exam-ined the DNA methylation status of MADS78 andMADS79. McrBC (an endonuclease that cleaves DNAcontaining methylcytosine) assay indicated that pro-moter regions ofMADS78 and 79 are methylated in theleaf tissue and developing seed (48 HAF; SupplementalFig. S12, A and B). Furthermore, bisulfite sequencingconfirmed methylation in the promoter region forMADS78 and the gene body forMADS79 in developingseeds (24, 48, 72, and 96 HAF; Supplemental Fig. S12C).No significant change in CG, CHG, or CHH methyla-tion levels for either gene were detected in developingseeds (Supplemental Fig. S12C). Thus, the methylationstatus did not associate with changes in the gene ex-pression profile of MADS78 and MADS79 (Fig. 1),suggesting that DNA methylation at the sites assayedis unlikely to play a direct regulatory role for thesetwo genes.

The FIS-PRC2 complex regulates imprinting by in-stalling histone methylation marks on the target genes(Folsom et al., 2014). In this framework, we asked ifFie1, an essential component of the FIS-PRC2 complex,is a potential regulator of MADS78 and MADS79. Thetranscript abundance of Fie1 was negligible at 48 HAFbut accumulated to significant levels by 72 HAF(Fig. 7A). Conversely, the transcript levels of MADS78and MADS79 were higher at 48 HAF and declinedconsiderably by 72 HAF in normally developing seeds(Fig. 7A). We next examined the expression of Fie1 inMADS78 and MADS79 mutant lines. Fie1 exhibitedlower transcript abundance at 72 HAF in the OE linescompared with the wild type (Fig. 7B). The singleknockout mutants of MADS78 and MADS79 accumu-lated more Fie1 transcripts at 48 HAF relative to thewild type (Fig. 7B). Moreover, in OE Fie1 mutants, thetranscripts of both MADS78 and MADS79 were notdetected at 48 HAF, when otherwise they were abun-dantly expressed in wild-type seeds (Fig. 7C). Also,expression ofMADS78 andMADS79was up-regulatedat 72 HAF in knockout mutants of Fie1 (Fig. 7D). Theexpression analysis of Fie1 in MADS78 and MADS79mutants and expression of MADS78 and MADS79in Fie1 mutants establishes an inverse transcriptionalassociation between MADS78 and MADS79 and Fie1.

Table 1. Physiological and morphometric measurements of OE lines and single knockout mutants relative to the wild type

Spikelet fertility, 1,000-grain weight, total tiller count, and total seeds per plant were measured at the whole-plant level (n 5 25–50 plants). Area,perimeter, length, width, and length-to-width ratio were measured for individual seeds (n 5 700–1,000). Tukey’s test was used for comparing thewild type and mutant lines: *, P , 0.05 and **, P , 0.001.

Plant Line 1,000-Grain Weight (g) Tiller Count Area (mm2) Perimeter (mm) Length (mm) Width (mm) Length-to-Width Ratio

Wild type 21.60 6 0.8 8.11 6 1.0 12.13 6 0.8 13.81 6 0.5 5.05 6 0.2 3.04 6 0.1 1.65 6 0.1OE78-4 19.09 6 2.0* 7.09 6 0.7 11.57 6 0.9** 13.58 6 0.6** 5.01 6 0.2* 2.90 6 0.1** 1.72 6 0.1**OE78-8 19.22 6 1.4* 6.44 6 1.0* 13.00 6 1.1** 14.42 6 0.6** 5.36 6 0.2** 3.05 6 0.1* 1.75 6 0.1**OE79-3 20.18 6 1.5 7.44 6 1.0 11.34 6 1.1** 13.44 6 1.0** 4.97 6 0.3** 2.87 6 0.2** 1.73 6 0.1**OE79-4 19.65 6 1.3 7.10 6 1.2 11.97 6 1.0 14.11 6 0.6** 5.38 6 0.2** 2.79 6 0.1** 1.93 6 0.1**mads78_17 19.17 6 1.7** 6.63 6 1.0 12.00 6 1.0* 13.66 6 0.7* 4.97 6 0.3* 3.05 6 0.01 1.62 6 0.1mads78_19 21.95 6 1.2 6.90 6 1.3 12.38 6 0.8 13.99 6 0.5* 5.14 6 0.2** 3.04 6 0.01 1.69 6 0.1**mads79_23 21.74 6 0.8 7.90 6 1.2 12.45 6 1.0* 14.01 6 0.6* 5.17 6 0.2** 3.06 6 0.01 1.68 6 0.1**mads79_24 19.40 6 0.5** 7.80 6 0.8 10.92 6 1.1** 13.05 6 0.7** 4.79 6 0.3** 2.90 6 0.01** 1.65 6 0.1


Paul et al.













Collectively, our data suggest that MADS78 andMADS79 regulation is likely linked to the abundanceof Fie1 and potentially mediated by the FIS-PRC2complex.

Misregulation of MADS78 and/or 79 DisturbsAuxin Homeostasis

To further explore the basis of the observed pheno-types with respect to misregulated endosperm cel-lularization, we performed transcriptome analysisof developing seed from mutants (48 and 72 HAF;Supplemental Table S3). We searched the tran-scriptome data set for genes that were (1) preferen-tially activated during the syncytial stage (48 HAF)and down-regulated with the onset of cellulariza-tion (72 HAF; syncytial-associated genes) and (2)suppressed during the syncytial stage but activa-ted with the progression of endosperm cellulari-zation, referred to here as cellularization-associatedgenes (Folsom et al., 2014; Chen et al., 2016). As ex-pected, OE mutants exhibiting delayed cellularizationhad higher transcript abundance for syncytium-associated genes at 48 and/or 72 HAF (SupplementalTable S3). Conversely, single knockout mutants withprecocious cellularization exhibited increased tran-script abundance of cellularization-associated genes

at 72 HAF relative to the wild type (SupplementalTable S4).Auxin (indole-3-acetic acid [IAA]) plays an important

role in the initiation of endosperm development and itsmaintenance during the proliferative phase in Arabi-dopsis (Figueiredo et al., 2015; Batista, 2019). Auxinsynthesized in endosperm is exported to integumentsvia the ABCB-type transporter, P-GLYCOPROTEIN10(PGP10; Figueiredo et al., 2016). Auxin blocks the PRC2complex and thus facilitates cell elongation and differ-entiation. Seeds deficient in AGL62 have reduced auxinexport due to suppressed PGP10, which releases PRC2suppression. The activated PRC2 in turn inhibits cellelongation and differentiation leading to precociouscellularization (Figueiredo et al., 2016; Batista et al.,2019a). Since MADS79 is one of the orthologs ofAGL62 and knocking out MADS78 or MADS79 accel-erates cellularization, we explored the link betweenauxin homeostasis and MADS78 and MADS79 in ourtranscriptome data. For this, we combined the tran-scriptome analysis with a search for potential targetgenes of MADS78 and/or MADS79 (SupplementalTable S5). The CArG box motif is a signature bindingmotif for MADS box TFs (both type I and II TFs;Batista et al., 2019b). The rice ABCB-type transporter,ABCB4, which carries a CArG box-like DNA-bindingmotif in its promoter, showed lower transcript abun-dance in the absence ofMADS78 orMADS79 compared

Figure 5. Single knockout mutants of MADS78and MADS79 show precocious cellularization.Representative cross sections of developing seedsare shown at 72, 84, and 96 HAF in singleknockout mutants of MADS78 and MADS79compared with the wild type (WT). The red ar-rowhead points to the central vacuole region (cv).ed, Endosperm. Bar 5 0.2 mm.

Figure 6. Double knockout mutation is lethal. A, Representative mature panicle images of T0 homozygous double knockoutmutants comparedwith the wild type (WT).mads78-79_5 andmads78-79_9 are derived from sgRNA-1, whilemads78-79_3 andmads78-79_4 correspond to sgRNA-2 (Supplemental Fig. S11). Bar 5 2 cm. B, Representative mature seed images of theT0 homozygous double knockout mutants with husk. Bar5 1 cm. Images shown in A and B were digitally extracted and scaledfor comparison.













with the wild type at 72 HAF (Fig. 8A). Likewise, otherABCB-type transporters, such asABCB11,ABCB14, andABCB22 (orthologs of Arabidopsis PGP10), also hadlower transcript levels in single knockout mutants,while expression in OE lines was unaltered (Fig. 8A;Supplemental Table S5). Genes responsible for main-taining auxin levels either by oxidation (DIOXYGEN-ASE FOR AUXIN OXIDATION [DAO]; Zhao et al.,2013) or amino acid conjugation (GRETCHENHAGEN3 [GH3]; Jain and Khurana, 2009) also showedlower transcript abundance in single knockouts at 72HAF (Fig. 8A; Supplemental Table S5). Furthermore,auxin signaling genes (IAA10, IAA19, and IAA24)exhibited lower transcript abundance in single knock-out mutants at 72 HAF (Fig. 8A; Supplemental TableS5). Previous work has shown that ABCB4/11/14/22,DAO, GH3s, and IAAs are up-regulated in developingrice seeds at 24 HAF (Jain and Khurana, 2009; Uchiumiand Okamoto, 2010; French et al., 2014; Basunia andNonhebel, 2019). TRYPTOPHAN AMINOTRANSFER-ASE-RELATED PROTEIN2 (TAR2), the auxin biosyn-thesis gene that is preferentially expressed prior tocellularization (Abu-Zaitoon et al., 2012), exhibitedlower transcript levels in MADS78 and MADS79 mu-tants, while YUCCA7 was unaltered in the MADS78mutant and showed higher transcript abundance inthe MADS79 mutant (Fig. 8A). On the other hand,auxin biosynthesis genes (YUCCA12 and TAR1) pref-erentially expressed in postcellularization seeds (Abu-Zaitoon et al., 2012; Basunia and Nonhebel, 2019)exhibited higher transcript abundance in singleknockouts at 72 HAF (Fig. 8A; Supplemental TableS5). YUCCA12 and TAR1 promoters have CArG box-binding motifs (Supplemental Table S5). Thesechanges in transcript abundance and the presenceof CArG box-binding motifs in promoters of auxinbiosynthesis, transporters, and signaling-related genes(Supplemental Table S5) suggest a regulatory role for

MADS78 and MADS79 in maintaining auxin levels indeveloping rice seeds (Fig. 8).

Misregulation of MADS78 and/or 79 AltersCarbon Metabolism

We observed a chalky grain appearance for matureseeds of OE lines and knockout mutants (Fig. 9A).Further investigation via scanning electron microscopy(SEM) revealed structural abnormalities in the starchgranules of the mutants. In contrast to densely packedpolygonal starch granules in the wild type, mutantspossessed loosely packed spherical starch granulesresulting in larger openings between granules (Fig. 9A).Quantification of total starch content from mature seedas well as developing seed (10 DAF) did not show asignificant difference between themutants and the wildtype (Supplemental Fig. S13A). However, we observedhigher total protein content for developing seed in OEand single knockout mutants relative to wild-typeseeds (Supplemental Fig. S13B). Abnormal starchgranule packaging and higher total protein contentare likely related to altered carbon and nitrogen me-tabolism. To investigate the metabolic status in thedeveloping seeds (10 DAF), we performed metaboliteprofiling by gas chromatography (GC) coupled to massspectrometry. Alterations in sugar metabolism wereobserved as evidenced by significant elevations in Glc,Fru, andmyoinositol in OE lines and knockout mutantsrelative to the wild type (Supplemental Table S6).Taken together, our results suggest impaired metabo-lism of carbohydrates in the developing seeds of themutant lines.

Genes involved in starch biosynthesis exhibited dif-ferential transcript abundance between the genotypes(Supplemental Table S7). We validated the expressionof a selected subset of these genes by RT-qPCR (Fig. 9B).In this context, the seed-specific isoform of Suc synthase(Susy3), the two subunits ofADP-Glc pyrophosphorylase

Figure 7. FIS-PRC2 potentially regulates MADS78and MADS79. A, RT-qPCR analysis of MADS78,MADS79, and Fie1 at 48 and 72 HAF in the wildtype (WT). Expression levels at 48 HAF were con-sidered as baseline for each gene. B, RT-qPCRanalysis of Fie1 in OE and single knockout mu-tants forMADS78 andMADS79 at 72 and 48 HAF,respectively. C and D, RT-qPCR analysis ofMADS78 andMADS79 inOE and single knockoutmutants for Fie1 at 48 and 72 HAF, respectively.For B to D, wild type expression was used as abaseline. For statistical analysis, the Holm-Sidakmethodwas used: ***, P, 0.001 and **, P, 0.01.Error bars indicate SD (n 5 6; two biological andthree technical replicates).


Paul et al.















(AGPS2b and AGPL2), and its transporter BRITTLE1showed higher transcript abundance in knockout mu-tants at 72 HAF, while the transcript levels were unal-tered in OE lines (Fig. 9B; Supplemental Table S7).The granule-bound starch synthase (GBSSI), whichis involved in amylose biosynthesis (Waxy), and itsupstream regulators, NF-YB1, exhibited higher tran-script levels in knockout mutants at 72 HAF (Fig. 9B;Supplemental Table S7). One of the amylopectin biosyn-thesis genes (starch synthase [SSIIa]), cytosolic pyruvate

orthophosphate dikinase (PPDK), and the known reg-ulators for seed storage proteins, rice basic Leu zipper(RISBZ1/bZIP58) and rice prolamin box binding factor(RPBF), were also altered. These genes showed highertranscript abundance in the developing seeds (72HAF) of OE lines and knockout mutants (Fig. 9B;Supplemental Table S7). Collectively, our results in-dicate that misregulation of MADS78 and MADS79during early seed development results in compro-mised seed quality.

DISCUSSION

Functional characterization ofMADS78 andMADS79revealed six key findings: (1) misregulation of the genesby either OE or knockout leads to delayed or precociousendosperm cellularization, respectively; (2) the doubleknockout mutant is not viable, thereby suggesting anindispensable role for MADS78 and MADS79 in riceseed development; (3) MADS78 and MADS79 phys-ically interact with MADS89, which enhances theirspecificity to localize in the nucleus; (4) MADS78 andMADS79 are potentially regulated by the FIS-PRC2complex; (5) MADS78 and MADS79, auxin, andFIS-PRC2 are important regulators of the timelytransition of endosperm from syncytium to cellula-rization; and (6) misregulation of MADS78 and/orMADS79 alters grain quality and metabolism.Functional characterization ofMADS78 andMADS79

indicates that both genes are involved in regulating therate of endosperm cellularization. Seeds deficient ineither of the two genes exhibit slightly precocious cel-lularization (Fig. 5), and overabundance of either generesults in delayed cellularization (Fig. 4D). Knockoutand OE lines of both genes do not arrest the initiationof cellularization but likely determine the duration re-quired to complete cellularization. Seeds that carrymutations in both MADS78 and MADS79 are nonvia-ble, indicating that at least one of these two genes isrequired for producing viable rice seeds (Fig. 6). Thisalso suggests that MADS78 and MADS79 have partialfunctional redundancy. OE of MADS78 has a greaterpenalty on spikelet fertility compared with MADS79(Fig. 4C). Although MADS78 and MADS79 do not di-merize with each other, their nuclear localization isenhanced when they heterodimerize with MADS89(Fig. 3). We found gene expression of Fie1, a riceFIS-PRC2 member, to be negatively associated with theabundance of MADS78 and MADS79 transcripts inknockout and OE lines, suggesting a possible link be-tween MADS78, MADS79, and Fie1 in regulating therate of endosperm cellularization (Fig. 7).

MADS89 Enhances Nuclear Localization of MADS78and MADS79

Both MADS78 and MADS79 belong to the Ma sub-clade and heterodimerized with two Mg proteins,MADS87 andMADS89. Furthermore, RNA-seq analysis

Figure 8. Misregulation of MADS78 and/or MADS79 disturbs auxinhomeostasis. A, Heat map based on normalized read counts for genesinvolved in auxin homeostasis in the wild type (WT) and singleknockout mutants (mads78 andmads79) in developing seeds (72 HAF).The color key represents a range of normalized values. B,MADS78 andMADS79, auxin, and FIS-PRC2 collectively regulate the timely initia-tion of endosperm cellularization. MADS78 and MADS79 regulatesome of the auxin biosynthesis (YUCCAs), signaling (IAAs and GH3s),and transport (ABCB-type transporters) related genes in developingendosperm. In normally developing syncytial stage seeds, auxin isexported from developing endosperm to the seed coat and blocksFIS-PRC2, thus ensuring normal endosperm proliferation and seed coatcell elongation. Seeds deficient inMADS78 andMADS79 have reducedauxin export, thus releasing auxin-mediated FIS-PRC2 suppression. Theactivated FIS-PRC2 blocks cell elongation and differentiation, whichcauses cellularization. The accelerated cellularization triggers the ac-tivation of auxin biosynthesis genes YUCCA12 and TAR1 that maypromote endoreduplication, differentiation, and starch deposition dur-ing later stages of seed development. ? indicates additional, unchar-acterized factors that might participate in this model.








revealed that both MADS87 and MADS89 coexpressedwith MADS78 and MADS79 (Supplemental Table S2).No homodimers were observed for either MADS78 orMADS79. In Arabidopsis, MADS box partnerships arelimited to subclass interactions, as interclass interac-tions (e.g. between type I and type II MADS) have notbeen reported (de Folter et al., 2005). Furthermore, in-teractions within a specific subclade of type I proteins(Ma, Mb, and Mg) are rare, and members of a subclademainly interact with a protein corresponding to anothersubclade. Ma proteins dimerize preferentially withmembers of Mb and Mg, and only a few interactionsexist between Mb and Mg. Thus, it is proposed thatMa proteins are central components stabilizing higherorder protein complexes (Immink et al., 2009). Thenecessity of at least one of MADS78 and MADS79being functional is indicated by the complete sterility ofdouble knockout mutants. Physical interactions of bothMADS78 and MADS79 with MADS89 suggest partialfunctional redundancy. One possible role for this in-teraction could be to enhance the nuclear localization ofMADS78 and MADS79 in the presence of MADS89.

Both MADS78 and MADS79 lack a clear NLS(Supplemental Table S2). The retargeting experimentsuggests that MADS78 and MADS79 require MADS89and not MADS87 to increase the specificity of their lo-calization to the nucleus (Fig. 6; Supplemental Fig. S4).The in silico analysis indicates that MADS89 carries anNLS on its N terminus, whereas MADS87 lacks an NLS(Supplemental Table S2). Likewise, in Arabidopsis,AGL62 (an ortholog of MADS79) heterodimerizes withAGL80 and AGL61, which are orthologs of MADS89and MADS78, respectively (de Folter et al., 2005).Both AGL80 and AGL61 play crucial roles in the initi-ation of endosperm development (de Folter et al., 2005;Portereiko et al., 2006; Steffen et al., 2008).

Misregulation of MADS78 and MADS79 Affects EarlySeed Development

MADS78 and MADS79 are preferentially expressedduring the syncytial stage of endosperm develop-ment, and transcript levels decline as the endosperm

Figure 9. Misregulation of MADS78and MADS79 causes impairment ofstarch granules. A, Representative im-ages of mature seeds from MADS78and MADS79 mutants (left column;bar 5 1 cm) and their cross sections(right column; bar 5 20 mm) usingSEM. Images shown were digitallyextracted and scaled for comparison. B,RT-qPCR analysis of selected genes in-volved in carbon metabolism at 72 HAFin OE and knockout mutants relative tothe wild type (WT). The genes were se-lected based on RNA-seq analysis. Forstatistical analysis, Student’s t test wasused: ***, P, 0.001; **, P, 0.01; and*, P, 0.05. Error bars indicate SD (n5 6;two biological and three technicalreplicates).


Paul et al.







cellularizes. Overexpressing either gene prevents thedecline in their transcript pools at 72 HAF (Fig. 4A).Higher abundance of MADS78 or MADS79 in OE linesis concomitant with delayed cellularization in seedsthat reach maturity. Seeds from wild-type plants arecellularized by 96 HAF. It is noteworthy that MADS78and MADS79 OE seeds have a range of phenotypes.While some seeds reach maturity, others can be scoredas abnormal as early as 48 HAF (before cellularization;Supplemental Fig. S9). These abnormal seeds are ran-domly distributed on the panicle. The development ofindividual seeds is partially independent of other seeds.Therefore, we hypothesize that certain developingseeds experience relatively higher syncytial activitythan others and thus fail to cope with this abnormaldevelopmental pace, eventually resulting in seedabortion. Seeds that abort during the early develop-mental stage contribute to the increased sterility ofmature panicles. On the other hand, some seeds endurethe abnormal developmental progression but get pe-nalized with respect to delayed endosperm cellulari-zation. Delayed cellularization in a subset of OE seedscould be due to (1) delayed transition from syncytiumto cellularization of the endosperm and/or (2) reducedrate of cellularization because of overabundance ofMADS78 andMADS79. Knockout mutants ofMADS78orMADS79 show precocious cellularization but do notimpact seed viability (Fig. 5; Supplemental Fig. S10,D and E). These observations are consistent with thereported role of AGL62 (an ortholog of MADS79 inArabidopsis) in endosperm cellularization (Kang et al.,2008; Bemer et al., 2010). agl62 mutant seeds showprecocious cellularization, thus establishing its rolein suppressing endosperm cellularization (Kang et al.,2008). Plants with mutations in both genes do notproduce viable seed, indicating that at least one of thesetwo genes should have a functional allele to produceviable seeds in rice (Fig. 6; Supplemental Fig. S11).

FIS-PRC2 Potentially Regulates MADS78 and MADS79

Normal endosperm development is tightly con-trolled by genome dosage, as paternal excess leads toendosperm overproliferation, while increase in mater-nal dosage results in precocious endosperm cellulari-zation (Scott et al., 1998; Stoute et al., 2012). FIS-PRC2is one of the key components that aids in maintaininggenome dosage by regulating seed-specific type IMADS box genes (Folsom et al., 2014). In this context,Fie1, which itself is regulated by repressive histonemethylation (H3K9me2), regulates type I MADS boxgenes (MADS82, MADS87, and AGL36) by H3K27me3histone methylation in rice (Folsom et al., 2014). Theexpression of type I MADS box genes MADS78 andMADS79 is negatively associated with Fie1 (Fig. 7).Overexpressing either MADS78 or MADS79 prolongsthe syncytium stage, thereby delaying the inductionof Fie1 expression (Fig. 7B). On the other hand, insingle knockout mutants of MADS78 and MADS79,

Fie1 transcripts accumulate at 48 HAF while remainingabsent in the wild type (Fig. 7B). Furthermore, expres-sion ofMADS78 andMADS79 is repressed at 48HAF inthe mutant lines overexpressing Fie1, whereas theirexpression is temporally extended in Fie1 knockoutmutants (Fig. 7, C and D). Our results are consistentwith previous work that reported an up-regulation ofAGL62 (an ortholog of MADS79) in PcG mutants ofArabidopsis (Kang et al., 2008).

Interplay between Auxin, Type I MADS Box TFs, andFIS-PRC2 Ensures Timely Endosperm Cellularization

We detected CArG box-binding motifs in the pro-moter of an auxin transporter, ABCB4 (SupplementalTable S5), which exhibits lower transcript abundance inknockout mutants (Fig. 8A). Moreover, auxin biosyn-thesis (TAR2) and signaling (IAAs and GH3s) genes,which are preferentially expressed before cellulariza-tion (Abu-Zaitoon et al., 2012), also showed lowertranscript abundance in knockout mutants, resulting inaltered auxin homeostasis (Fig. 8). This could lead todecreased auxin export, which relieves FIS-PRC2 sup-pression in the seed coat (Figueiredo et al., 2016), asevidenced by increased transcript abundance of Fie1 inknockout mutants (Fig. 7B). The activated FIS-PRC2likely blocks cell elongation and differentiation, whichinduces precocious cellularization (Batista et al., 2019a).Early cellularization in the absence of MADS78 orMADS79 triggers activation of the CArG box-containingauxin biosynthesis genes YUCCA12 and TAR1 (Fig. 8A)that may have a role in subsequent endoreduplication,differentiation, and starch deposition during laterstages of seed development (Abu-Zaitoon et al., 2012;Basunia and Nonhebel, 2019). Detection of CArGbox-binding motifs in promoters of auxin biosyn-thesis enzyme-, transporter-, and signaling-relatedgenes further supports the regulatory role of type IMADS box TFs in maintaining auxin levels and/ordistribution (Supplemental Table S5). Collectively,our findings present evidence that type I MADS boxTFs (MADS78 and MADS79 and potentially othertype I genes), in conjunction with auxin homeostasisand Fie1, regulate the timely transition of endospermfrom syncytial stage to cellularization, which is neces-sary for normal seed development (Fig. 8B).

Misregulation of Early Seed Development Alters GrainQuality Traits

Timely transition of endosperm from syncytial tocellularization ensures optimal nutritional quality ofthe seed, which is also dependent upon successful im-port of photoassimilates from source tissues (Aguirreet al., 2018).Misregulation ofMADS78 and/orMADS79,associated with mistimed endosperm cellularization,also causes structural abnormalities in kernel starch(Fig. 9A). The starch biosynthesis genes were either












unchanged or down-regulated in OE lines and up-regulated in single knockout mutants (Fig. 9B;Supplemental Table S7). This alteration implied thatOE mutants exhibiting delayed cellularization post-pone the transition of developing seed from the pre-storage phase to the storage phase, while singleknockout mutants with precocious cellularizationreach the storage phase earlier. The transcript levelchanges for starch biosynthesis genes did not cor-roborate with starch quantity; however, abnormalstarch granule structures were observed (Fig. 9A).This is not entirely unexpected, because the starchgranule structure is determined not only by starchcontent but also by protein content, protein-starchinteractions, amylose-lipid complexes, and shortand long chains of amylopectin, which collectivelydefine cooking and milling qualities of rice (Singhet al., 2007; Chen et al., 2012). The discrepancy be-tween starch quantity and related gene expres-sion is linked to the cytosolic (cy) isoform of PPDK,which is preferentially expressed in endosperm,especially during syncytial/endosperm cellulariza-tion, and down-regulated during subsequent stagesby posttranslational mechanisms (Kang et al., 2005;Chastain et al., 2006). cyPPDK catalyzes a reversiblereaction between pyruvate and phosphoenolpyr-uvate, thereby modulating carbon metabolism in fa-vor of amino acid or lipid biosynthesis (Aoyagi andBassham, 1984; Kang et al., 2005). In our study,cyPPDK showed higher transcript abundance in sin-gle knockout mutants (Fig. 9B; Supplemental TableS7). These results indicate that carbon influx is pos-sibly directed toward either amino acid or lipid bio-synthesis. Higher transcript abundance for SSIIa,RISBZ1/bZIP58, and RPBF was detected for both OElines and knockout mutants (Fig. 9B; SupplementalTable S7). Since normal starch deposition and endo-sperm development are impaired in both OE linesand knockout mutants, it is plausible that SSIIaand potentially other genes are misregulated (at thetranscript level) as a consequence of abnormal seeddevelopment triggered by overabundance or defi-ciency of MADS78 and MADS79. It is important tonote that in normally developing seeds,MADS78 andMADS79 transcript abundance declines as endo-sperm initiates starch biosynthesis in nuclei that havecellularized.

CONCLUSION

MADS78 and MADS79 play an important role inregulating rice seed size and shape. Their misregulationis associated with temporal defects in early seed de-velopment and abnormal starch deposition. Since riceyield and milling quality are significantly impacted byunfavorable environments, such as drought and tem-perature stress, it will be useful to explore if MADS78and MADS79 could be mediating the stress sensitivityof rice yield and quality traits in future research.

MATERIALS AND METHODS

Protein Sequence Alignment and Promoter Analysis

Protein sequences for MADS78 (LOC_Os09g02830) and MADS79(LOC_Os01g74440) were downloaded from RGAP7. The SMART database(Letunic et al., 2012) was used to predict domains. The T-COFFEE programwasused for aligning protein sequences of MADS78 and MADS79 (Di Tommasoet al., 2011). For promoter analysis, the regions 2 kb upstream of the tran-scription start sites of both MADS78 and MADS79 were downloaded fromRGAP7 and analyzed using the PlantCARE database (Lescot et al., 2002).

Generation of Transgenic Plants

For generating overexpression constructs, full-length CDS corresponding toMADS78 andMADS79were amplified. OEmutants of Fie1 (LOC_Os08g04290)were considered from Folsom et al. (2014). The amplified products were clonedinto pENTR/D-TOPO entry vector (Invitrogen). PCR products were movedfrom entry to destination vector (Ubi-NC1300) using LR clonase (Invitrogen).The final constructs were transformed into Agrobacterium tumefaciens strainEHA105. Calli from rice (Oryza sativa ‘Kitaake’) were transformed as describedby Cheng et al. (1998). Hygromycin was used as a selective agent to screen thetransformed calli. To generate CRISPR-Cas9 mutants, sgRNAs were designedusing CRISPR-P 1.0 (http://crispr.hzau.edu.cn/CRISPR/) to specifically targeteither MADS78 or MADS79 and Fie1 (Lei et al., 2014). Considering the highpercentage of sequence similarity between MADS78 and MADS79, sgRNAswere designed to target nonconserved genic regions. For targetingMADS78, ansgRNA that targeted a region downstream of the coiled-coil domain was se-lected. An sgRNA targeting a region between the MADS and coiled-coil do-mains was designed for MADS79 (Supplemental Fig. S10A). A list of potentialoff-targets corresponding to genic regions (exonic or intronic) for each sgRNA isprovided in Supplemental Table S8. To create double knockout constructs, wedesigned two sgRNAs (sgRNA-1 and -2) targeting bothMADS78 andMADS79in their conserved MADS and coiled-coil domains (Supplemental Fig. S11A).With the aim of introducing point mutations in both genes, separate transfor-mations for both sgRNAs were performed. The primer sequences used in thestudy are listed in Supplemental Table S9.

CRISPR/Cas9 constructs were developed following the protocols describedby Lowder et al. (2015). pYPQ141C was used to clone all sgRNAs, which werethen transferred to a destination vector (pANIC6B) along with a vector con-taining Cas9 (pYPQ167) using LR clonase. The destination vector was trans-formed into A. tumefaciens, which was further used to infect calli. For singleknockout mutants, T1 segregates carrying mutations and lacking Cas9 (con-firmed by GUS screening assay) were considered for downstream analysis(Lowder et al., 2015). T0, T1, and T2 plants were screened for mutations bySanger sequencing. T2 or later generations of mutant plant lines were used inthe study. For experiments, mature rice (‘Kitaake’) seeds were manuallydehusked and sterilized with bleach (40%, v/v) and water. The seeds weregerminated in the dark on one-half-strength Murashige and Skoog medium.The germinated seedlings were transplanted in soil amid controlled greenhouseconditions: 16 h of light and 8 h of dark at 28°C 6 1°C and 23°C 6 1°C, re-spectively, and a relative humidity of 55% to 60% (Sandhu et al., 2019).

Genomic DNA and RNA Extraction, RT-qPCR

To screen formutations in the single and double knockoutmutants, genomicDNAwas isolated from leavesusing the Sucmethod (Berendzen et al., 2005). Theregion of interest was amplified using Kapa3G Plant PCR Kits (Kapa Biosys-tems) according to the manufacturer’s protocol. The amplicon was then se-quenced for genotyping. Total RNA was extracted from the following tissues:seedling, leaves, mature pollen, unfertilized ovary, and developing seed (24, 48,72, and 96 HAF), using the RNeasy Plant mini kit (Qiagen). RNA was subse-quently DNase treated. Early seed developmental stages corresponding to24, 48, 72, and 96 HAF (normally developing seed tissues: category B;Supplemental Fig. S9), as well as unfertilized ovaries, were collected as de-scribed (Folsom et al., 2014). Spikelets were marked at the time of fertilization,which is evident by the opening of a floret, to track the precise developmentalstage. Mature pollen grains were collected as described (Paul et al., 2016, 2017)withminor adjustments. After anthesis, the protruded anthers were collected in500 mL of germination solution (2 mM boric acid, 2 mM calcium nitrate, 2 mM

magnesium sulfate, and 1 mM potassium nitrate). The collected anthers werevortexed, filtered through muslin cloth, and centrifuged for 1 min at 1,000g.


Paul et al.







http://crispr.hzau.edu.cn/CRISPR/







The pelleted pollen grainswerewashedmultiple times in 200mL of germinationsolution to remove unwanted material. The collected pollen grains were im-mersed in liquid nitrogen and stored at 280°C.

One microgram of total RNA was used for cDNA synthesis using theSuperScript VILO cDNA synthesis kit (Invitrogen). The RT-qPCR (10 mL) com-prising gene-specific primers, SYBR Green Master Mix (Bio-Rad), and templatewas conductedusing the Lightcycler 480Real-TimePCRSystem (Roche). Thermalcycling conditions were used as described by Fragkostefanakis et al. (2015). As anendogenous control, ubiquitin (UBQ5) genes were used as reference genes (Jainet al., 2006). Data were analyzed using standard methods (Livak and Schmittgen,2001). For RT-qPCR assays, a minimum of two independent biological replicatesand three technical replicates were conducted.

Pollen Viability Assay, Agronomic andMorphometric Measurements

For pollen viability assays, mature pollen was collected and subjected tostaining with Alexander dye (Alexander, 1980) and Lugol’s solution (I2/KI).Pollen number was determined using a Fuchs-Rosenthal cell counter(Fragkostefanakis et al., 2016). Three independent biological replicates wereperformed for the pollen viability assay. For determining grain morphology,seeds were harvested at maturity, and only fully developed seeds were scoredto calculate percentage of spikelet fertility. The fully developed seeds weredehusked using a rice husker (Kett TR-130). The dehusked seeds were scannedusing an Epson Expression 12000 XL scanner andwere analyzed by SmartGrainsoftware to obtain seed size measurements (Tanabata et al., 2012). For devel-oping seed analysis, florets marked at the time of fertilization were tracked for4 d after fertilization (24, 48, 72, and 96 HAF). Each marked floret was openedand evaluated for seed abortion at the designated time. Images of the normaland aborted seeds were representative across all three replicates.

DNA Methylation Assay

GenomicDNAwas extracted from twobiological replicates of seeds at 24, 48,72, and 96 HAF using the Qiagen DNeasy Plant Mini Kit. Genomic DNA wasthen treated with sodium bisulfite (EZ DNAMethylation Kit; Zymo Research).Whole-genome bisulfite sequencing was performed using an Illumina high-throughput sequencing platform to generate 126-bp paired end reads. Afterfiltering the low-quality regions using Trimmomatic, v 0.32 (Bolger et al., 2014),reads were mapped to the reference genome (rice ssp. japonica MSU v7.0), andmethylated sites were called using the Bismark v0.16.3 pipeline (Krueger andAndrews, 2011). Methylation levels of sites corresponding to the genomic re-gions ofMADS78 andMADS79were plotted using R-studio. McrBC assay wasperformed as mentioned by Folsom et al. (2014).

Histochemical Analysis and RNA in SituHybridization Assay

Normally developing seed tissue (category B; Supplemental Fig. S9) corre-sponding to 72, 84, and 96 HAF was harvested and fixed in 1 mL of formal-dehyde (2%, v/v), acetic acid (5%, v/v), and ethanol (60% [v/v]; FAE solution)for 16 h, followed by a 70% (v/v) ethanol wash and stored overnight at 4°C. Thedehydration procedure was performed with an ethanol series (85%, 95%, and100% [v/v]) for 1 h each at room temperature. Samples were incubated inethanol:xylene (1:1) for 1 h, followed by xylene (100%) for 2 h, and thentransferred to 500mL of xylene and Paraplast tablets at 60°C. Sampleswere thentransferred and embedded in Paraplast. Ten-micrometer sections were madeusing a rotary microtome (Leica RM2125 RTS). RNA in situ hybridization wasperformed following the protocol described by Chen et al. (2016). Gene-specificfragments were amplified forMADS78 andMADS79. The respective ampliconswere cloned into pGEM-T Easy vector (Promega). RNA probes were generatedby in vitro transcription using the DIG RNA labeling kit (Roche). The RNAtranscripts by T7 and SP6 polymerase were applied to slides as sense and an-tisense probes, respectively. Images from cross sections and in situ hybridiza-tion were observed using a bright-field microscope (Leica DM-2500) andrepresent six independent biological replicates.

Localization and BiFC

For localization, the CDS regions ofMADS78 andMADS79were cloned intopGWB5 and pGWB6 vectors (obtained from TAIR), generating protein fusions

betweenGFP and their C andN termini, respectively. Empty vectorwas used asa control. For BiFC, full-length CDS of 11 rice type IMADS box genes (MADS70,MADS77, MADS81, MADS83, MADS84, MADS87, MADS88, MADS89,MADS90, MADS96, and MADS98) were amplified and cloned into eitherYFC43 or YFN43 containing the C- or N-terminal half of yellow fluorescentprotein (YFP), respectively. Gene selection for the interaction study was basedon their spatiotemporal coexpression with MADS78 and MADS79 (Chen et al.,2016). To rule out any false positives, we swapped the vectors for MADS78 andMADS79 with positive interactors (MADS87 and MADS89). Empty vector andbZIP76, an unrelated TF that is expressed during early seed development, wereused for controls. Also, we cotransformed the MADS78 and MADS79 GFPfusions (cloned in pGWB5 and pGWB6) with a nonfluorescent BiFC construct(YFN43), encoding a positive interaction partner (MADS87 or MADS89), to testthe translocation efficacy of each gene into the nucleus. For both localizationand BiFC experiments,Nicotiana benthamiana plants were hand infiltrated usingneedleless syringes containing suspensions of A. tumefaciens (EHA105 strain)transformed with each vector corresponding to a gene of interest. Two dayslater, fluorescence signals were observed using a confocal laser scanning mi-croscope (Nikon A1). The following conditions were used for imaging: GFP,excitation at 488 nm and emission at 500 to 550 nm; YFP, excitation at 514.5 nmand emission at 525 to 555 nm (pseudocolored green); chlorophyll, excitation at640.9 nm and emission at 663 to 738 nm. The images shown for localizationand BiFC experiments are representative of four independent experimentswith three to four biological replicates in each experiment. For in silico NLSprediction, we used cNLS mapper (http://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi; Kosugi et al., 2009).

SEM

For SEM,mature rice grainswere transversely cut, placed on conductive tapeattached to aluminum SEM stubs, and vacuumdried in a sample drying oven at40°C for 1 week. Afterward, samples were sputter coated with chromium usinga Denton Vacuum Desk V sputter coater and imaged with the Hitachi-S4700Field-Emission SEM device at 503, 2003, and 2,0003 (Dhatt et al., 2019). Threeindependent biological replicates were used for the observations.

Metabolite Profiling and Quantification of Total Starch andProtein Content

Polar metabolites were extracted from 25 mg of developing seeds (10 DAF),and 50 mL of extract was dried and derivatized as described (Lisec et al., 2006).Five biological replicates were considered for metabolite profiling, which wasperformed using a 7200GC-QTOF system (Agilent) with anHP-5MSUI column(30 m, 0.25 mm, and 0.25 mm; Agilent). GC and ionization parameters followedthose described by Lisec et al. (2006). Mass spectrometry detection was per-formed in TOF mode with a 50-Hz scan rate. Peaks in the reference samplecontaining equal volumes of all samples were annotated with the metabolitesin the Fiehn GC/MS Metabolomics RTL Library (Agilent) by MassHunterUnknown Analysis (Agilent) with manual curation. Peak areas of each repre-sentative ion were quantified by MassHunter Quantitative Analysis (Agilent).Following background subtraction, peak area was normalized to that of theinternal standard (ribitol) and sample fresh weight to calculate relative me-tabolite levels with orbital unit. The parameters used for peak annotation andpeak area values are listed in Supplemental Table S10. Starch and total proteincontents were determined in the pellet of metabolite extraction as described(Gibon et al., 2009).

RNA-Seq Analysis

Isolationof totalRNAfromdeveloping seed (48 and72HAF)of thewild type,OE lines, and single knockout mutants was performed as described above. TheNuGen Universal Plus mRNA-Seq kit was used for preparing RNA-seq li-braries. The libraries were sequenced on a single-read, 75-bp high-output flowcell on the NextSeq 550 device. For the analysis, each RNA-seq read wastrimmed using Trimmomatic to ensure that it had a minimum length of 70 bpand an average quality score greater than 30 (Bolger et al., 2014). All trimmedshort reads were mapped to the rice genome (RGAP v7.0) using TopHat,allowing up to two base mismatches per read (Trapnell et al., 2009). Readsmapped to multiple locations were not considered. Numbers of reads in geneswere counted by the software tool HTSeq-count using corresponding rice gene







annotations and the union resolution mode (Supplemental Table S4; Anderset al., 2015).

Target Gene Analysis

FIMO (http://meme-suite.org/tools/fimo) was used to search for MADSbox-binding sites (Grant et al., 2011). Promoter sequences (22,000 bp upstream)downloaded from the MSU7 annotation (Kawahara et al., 2013) were scannedformatcheswith the CArG box-binding sequence (Batista et al., 2019b) obtainedfrom JASPAR 2018 (http://jaspar.genereg.net/; Khan et al., 2018).

Accession Numbers

Gene accession numbers are displayed in Supplemental Table S2.

Supplemental Data

The following supplemental materials are available.

Supplemental Figure S1. Protein sequence alignment for MADS78 andMADS79.

Supplemental Figure S2. In silico promoter analysis.

Supplemental Figure S3. Empty vector used as an additional control forBiFC assays.

Supplemental Figure S4. Confirmation of positive protein-proteininteractions.

Supplemental Figure S5. Localization of MADS78 and 79 in the absenceand presence of the positive interactors, MADS89 and 87.

Supplemental Figure S6. Representative pictures for overexpression mu-tants at the vegetative stage.

Supplemental Figure S7. Pollen viability assay.

Supplemental Figure S8. Mature sterile seeds from overexpression mu-tants of MADS78 and MADS79.

Supplemental Figure S9. Seed abortion analysis during early seeddevelopment.

Supplemental Figure S10. Single knockout mutants for MADS78 andMADS79.

Supplemental Figure S11. Double knockout mutations are lethal.

Supplemental Figure S12. DNA methylation analysis.

Supplemental Figure S13. Quantification of total starch and total proteincontent from developing and mature seeds.

Supplemental Table S1. RNA-seq-based normalized read counts forMADS78 and MADS79.

Supplemental Table S2. Normalized read counts for type I MADS boxgenes in developing rice seeds (48 and 72 HAF).

Supplemental Table S3. Summary of RNA-seq libraries.

Supplemental Table S4. Normalized read count for syncytial- andcellularization-associated genes.

Supplemental Table S5. Potential target genes for MADS78 and MADS79.

Supplemental Table S6. Levels of metabolites in the developing grains (10DAF) of MADS78 and MADS79 mutants.

Supplemental Table S7. Normalized read count for carbon metabolism-related genes in mutants and the wild type at 72 HAF.

Supplemental Table S8. List of potential off-targets (as per CRISPR-P 1.0)corresponding to genic region (exonic or intronic) for sgRNAs and theirnormalized read counts in single knockout mutants.

Supplemental Table S9. List of primers used in the study.

Supplemental Table S10. Parameters used for peak annotation and rowpeak area data for metabolite profiling.

ACKNOWLEDGMENTS

We thank Christian Elowsky, Jules Russ, and You Zhou from the MorrisonCore Research Facility, Center for Biotechnology, University of Nebraska-Lincoln, for help with microscopywork. RNA-seq data analysis was completedutilizing the Holland Computing Center of the University of Nebraska. Wethank Martha Rowe (Department of Agronomy and Horticulture, Universityof Nebraska-Lincoln) for help with the histochemical analysis.

Received July 29, 2019; accepted November 27, 2019; published December 5,2019.

LITERATURE CITED

Abu-Zaitoon YM, Bennett K, Normanly J, Nonhebel HM (2012) A largeincrease in IAA during development of rice grains correlates with theexpression of tryptophan aminotransferase OsTAR1 and a grain-specificYUCCA. Physiol Plant 146: 487–499

Aguirre M, Kiegle E, Leo G, Ezquer I (2018) Carbohydrate reserves andseed development: An overview. Plant Reprod 31: 263–290

Alexander MP (1980) A versatile stain for pollen fungi, yeast and bacteria.Stain Technol 55: 13–18

Anders S, Pyl PT, Huber W (2015) HTSeq: A Python framework to workwith high-throughput sequencing data. Bioinformatics 31: 166–169

Aoyagi K, Bassham JA (1984) Pyruvate orthophosphate dikinase of c(3)seeds and leaves as compared to the enzyme from maize. Plant Physiol75: 387–392

Arora R, Agarwal P, Ray S, Singh AK, Singh VP, Tyagi AK, Kapoor S(2007) MADS-box gene family in rice: Genome-wide identification, or-ganization and expression profiling during reproductive developmentand stress. BMC Genomics 8: 242

Basunia MA, Nonhebel HM (2019) Hormonal regulation of cereal endo-sperm development with a focus on rice (Oryza sativa). Funct Plant Biol46: 493–506

Batista RA (2019) Signalling mechanisms and epigenetic regulation of seeddevelopment in Arabidopsis thaliana. PhD thesis. Swedish University ofAgricultural Sciences, Uppsala, Switzerland

Batista RA, Figueiredo DD, Santos-González J, Köhler C (2019a) Auxinregulates endosperm cellularization in Arabidopsis. Genes Dev 33:466–476

Batista RA, Moreno-Romero J, Qiu Y, van Boven J, Santos-González J,Figueiredo DD, Köhler C (2019b) The MADS-box transcription factorPHERES1 controls imprinting in the endosperm by binding to domes-ticated transposons. eLife 8: 616698

Bemer M, Heijmans K, Airoldi C, Davies B, Angenent GC (2010) An atlasof type I MADS box gene expression during female gametophyte andseed development in Arabidopsis. Plant Physiol 154: 287–300

Berendzen K, Searle I, Ravenscroft D, Koncz C, Batschauer A, CouplandG, Somssich IE, Ülker B (2005) A rapid and versatile combined DNA/RNAextraction protocol and its application to the analysis of a novel DNA markerset polymorphic between Arabidopsis thaliana ecotypes Col-0 and Lands-berg erecta. Plant Methods 1: 4

Bjerkan KN, Hornslien KS, Johannessen IM, Krabberød AK, vanEkelenburg YS, Kalantarian M, Shirzadi R, Comai L, Brysting AK,Bramsiepe J, et al (2019) Genetic variation and temperature affects hy-brid barriers during interspecific hybridization. Plant J doi:10.1111/tpj.14523

Blackwell TK, Bowerman B, Preiss JR, Weintraub H (1994) Formation of amonomeric DNA binding domain by Skn-1 bZIP and homeodomainelements. Science 266: 621–628

Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: A flexible trimmer forIllumina sequence data. Bioinformatics 30: 2114–2120

Brown RC, Lemmon BE, Olsen OA (1996) Development of the endospermin rice (Oryza sativa L.): Cellularization. J Plant Res 109: 301–313

Chastain CJ, Heck JW, Colquhoun TA, Voge DG, Gu XY (2006) Post-translational regulation of pyruvate, orthophosphate dikinase in de-veloping rice (Oryza sativa) seeds. Planta 224: 924–934

Chaudhury AM, Koltunow A, Payne T, Luo M, Tucker MR, Dennis ES,Peacock WJ (2001) Control of early seed development. Annu Rev CellDev Biol 17: 677–699

Chaudhury AM, Ming L, Miller C, Craig S, Dennis ES, Peacock WJ (1997)Fertilization-independent seed development in Arabidopsis thaliana. ProcNatl Acad Sci USA 94: 4223–4228


Paul et al.



http://meme-suite.org/tools/fimo


























Chen C, Begcy K, Liu K, Folsom JJ, Wang Z, Zhang C, Walia H (2016)Heat stress yields a unique MADS box transcription factor in deter-mining seed size and thermal sensitivity. Plant Physiol 171: 606–622

Chen C, Li T, Zhu S, Liu Z, Shi Z, Zheng X, Chen R, Huang J, Shen Y, LuoS, et al (2018) Characterization of imprinted genes in rice reveals con-servation of regulation and imprinting with other plant species. PlantPhysiol 177: 1754–1771

Chen Y, Wang M, Ouwerkerk PBF (2012) Molecular and environmentalfactors determining grain quality in rice. Food Energy Secur 1: 111–132

Cheng X, Sardana RK, Altosaar I (1998) Rice transformation by Agro-bacterium infection. In C Cunningham, and AJR Porter, eds, Recombi-nant Proteins from Plants Methods in Biotechnology, Vol Vol 3.Humana Press, Totawa, NJ, pp 1–9

de Folter S, Immink RG, Kieffer M, Parenicová L, Henz SR, Weigel D,Busscher M, Kooiker M, Colombo L, Kater MM, et al (2005) Com-prehensive interaction map of the Arabidopsis MADS box transcriptionfactors. Plant Cell 17: 1424–1433

Dhatt BK, Abshire N, Paul P, Hasanthika K, Sandhu J, Zhang Q, Obata T,Walia H (2019) Metabolic dynamics of developing rice seeds under highnight-time temperature stress. Front Plant Sci 10: 1443

Dickinson CD, Evans RP, Nielsen NC (1988) RY repeats are conserved inthe 59-flanking regions of legume seed-protein genes. Nucleic Acids Res16: 371

Di Tommaso P, Moretti S, Xenarios I, Orobitg M, Montanyola A, ChangJM, Taly JF, Notredame C (2011) T-Coffee: A web server for the mul-tiple sequence alignment of protein and RNA sequences using structuralinformation and homology extension. Nucleic Acids Res 39: W13–W17

Figueiredo DD, Batista RA, Roszak PJ, Hennig L, Köhler C (2016) Auxinproduction in the endosperm drives seed coat development in Arabi-dopsis. eLife 5: 1–23

Figueiredo DD, Batista RA, Roszak PJ, Köhler C (2015) Auxin productioncouples endosperm development to fertilization. Nat Plants 1: 15184

Finnegan EJ, Peacock WJ, Dennis ES (1996) Reduced DNA methylation inArabidopsis thaliana results in abnormal plant development. Proc NatlAcad Sci USA 93: 8449–8454

Folsom JJ, Begcy K, Hao X, Wang D, Walia H (2014) Rice fertilization-Independent Endosperm1 regulates seed size under heat stress by con-trolling early endosperm development. Plant Physiol 165: 238–248

Fragkostefanakis S, Mesihovic A, Simm S, Paupière MJ, Hu Y, Paul P,Mishra SK, Tschiersch B, Theres K, Bovy A, et al (2016) HsfA2 controlsthe activity of developmentally and stress-regulated heat stress protec-tion mechanisms in tomato male reproductive tissues. Plant Physiol 170:2461–2477

Fragkostefanakis S, Simm S, Paul P, Bublak D, Scharf KD, Schleiff E(2015) Chaperone network composition in Solanum lycopersicum ex-plored by transcriptome profiling and microarray meta-analysis. PlantCell Environ 38: 693–709

French SR, Abu-Zaitoon Y, Uddin MM, Bennett K, Nonhebel HM (2014)Auxin and cell wall invertase related signaling during rice grain de-velopment. Plants (Basel) 3: 95–112

Gao Y, Xu H, Shen Y, Wang J (2013) Transcriptomic analysis of rice (Oryzasativa) endosperm using the RNA-Seq technique. Plant Mol Biol 81:363–378

Garcia D, Fitz Gerald JN, Berger F (2005) Maternal control of integumentcell elongation and zygotic control of endosperm growth are coordi-nated to determine seed size in Arabidopsis. Plant Cell 17: 52–60

Gehring M (2013) Genomic imprinting: Insights from plants. Annu RevGenet 47: 187–208

Genger RK, Kovac KA, Dennis ES, Peacock WJ, Finnegan EJ (1999)Multiple DNA methyltransferase genes in Arabidopsis thaliana. Plant MolBiol 41: 269–278

Gibon Y, Pyl ET, Sulpice R, Lunn JE, Höhne M, Günther M, Stitt M (2009)Adjustment of growth, starch turnover, protein content and centralmetabolism to a decrease of the carbon supply when Arabidopsis isgrown in very short photoperiods. Plant Cell Environ 32: 859–874

Grant CE, Bailey TL, Noble WS (2011) FIMO: Scanning for occurrences of agiven motif. Bioinformatics 27: 1017–1018

Grossniklaus U, Vielle-Calzada J, Hoeppner M, Gagliano W (1998) Ma-ternal control of embryogenesis by MEDEA, a polycomb group gene inArabidopsis. Science 280: 446–450

Haig D (2013) Kin conflict in seed development: An interdependent butfractious collective. Annu Rev Cell Dev Biol 29: 189–211

Huh JH, Bauer MJ, Hsieh TF, Fischer R (2007) Endosperm gene imprintingand seed development. Curr Opin Genet Dev 17: 480–485

Immink RG, Tonaco IA, de Folter S, Shchennikova A, van Dijk AD,Busscher-Lange J, Borst JW, Angenent GC (2009) SEPALLATA3: The‘glue’ for MADS box transcription factor complex formation. GenomeBiol 10: R24

Jain M, Khurana JP (2009) Transcript profiling reveals diverse roles ofauxin-responsive genes during reproductive development and abioticstress in rice. FEBS J 276: 3148–3162

Jain M, Nijhawan A, Tyagi AK, Khurana JP (2006) Validation of house-keeping genes as internal control for studying gene expression in rice byquantitative real-time PCR. Biochem Biophys Res Commun 345: 646–651

Jullien PE, Berger F (2010) Parental genome dosage imbalance deregulatesimprinting in Arabidopsis. PLoS Genet 6: e1000885

Kang HG, Park S, Matsuoka M, An G (2005) White-core endosperm flouryendosperm-4 in rice is generated by knockout mutations in the C-typepyruvate orthophosphate dikinase gene (OsPPDKB). Plant J 42: 901–911

Kang IH, Steffen JG, Portereiko MF, Lloyd A, Drews GN (2008) TheAGL62 MADS domain protein regulates cellularization during endo-sperm development in Arabidopsis. Plant Cell 20: 635–647

Kawahara Y, De La Bastide M, Hamilton JP, Kanamori H, McCombieWR, Ouyang S, Schwartz DC, Tanaka T, Wu J, Zhou S, et al (2013)Improvement of the Oryza sativa Nipponbare reference genome usingnext generation sequence and optical map data. Rice (N Y) 6: 4

Khan A, Fornes O, Stigliani A, Gheorghe M, Castro-Mondragon JA, vander Lee R, Bessy A, Chèneby J, Kulkarni SR, Tan G, et al (2018)JASPAR 2018: Update of the open-access database of transcription factorbinding profiles and its web framework. Nucleic Acids Res 46:D260–D266

Köhler C, Hennig L, Spillane C, Pien S, Gruissem W, Grossniklaus U(2003) The Polycomb-group protein MEDEA regulates seed develop-ment by controlling expression of the MADS-box gene PHERES1. GenesDev 17: 1540–1553

Köhler C, Wolff P, Spillane C (2012) Epigenetic mechanisms underlyinggenomic imprinting in plants. Annu Rev Plant Biol 63: 331–352

Kosugi S, Hasebe M, Tomita M, Yanagawa H (2009) Systematic identifi-cation of cell cycle-dependent yeast nucleocytoplasmic shuttling pro-teins by prediction of composite motifs. Proc Natl Acad Sci USA 106:10171–10176

Krueger F, Andrews SR (2011) Bismark: A flexible aligner and methylationcaller for Bisulfite-Seq applications. Bioinformatics 27: 1571–1572

Lei Y, Lu L, Liu HY, Li S, Xing F, Chen LL (2014) CRISPR-P: A web tool forsynthetic single-guide RNA design of CRISPR-system in plants. MolPlant 7: 1494–1496

Lescot M, Déhais P, Thijs G, Marchal K, Moreau Y, Van de Peer Y, RouzéP, Rombauts S (2002) PlantCARE, a database of plant cis-acting regu-latory elements and a portal to tools for in silico analysis of promotersequences. Nucleic Acids Res 30: 325–327

Letunic I, Doerks T, Bork P (2012) SMART 7: Recent updates to the proteindomain annotation resource. Nucleic Acids Res 40: D302–D305

Lisec J, Schauer N, Kopka J, Willmitzer L, Fernie AR (2006) Gas chro-matography mass spectrometry-based metabolite profiling in plants.Nat Protoc 1: 387–396

Livak KJ, Schmittgen TD (2001) Analysis of relative gene expression datausing real-time quantitative PCR and the 22D D CT method. Methods 25:402–408

Lowder LG, Zhang D, Baltes NJ, Paul JW III, Tang X, Zheng X, VoytasDF, Hsieh TF, Zhang Y, Qi Y (2015) A CRISPR/Cas9 toolbox for mul-tiplexed plant genome editing and transcriptional regulation. PlantPhysiol 169: 971–985

Luo M, Bilodeau P, Dennis ES, Peacock WJ, Chaudhury A (2000) Ex-pression and parent-of-origin effects for FIS2, MEA, and FIE in the en-dosperm and embryo of developing Arabidopsis seeds. Proc Natl AcadSci USA 97: 10637–10642

Luo M, Platten D, Chaudhury A, Peacock WJ, Dennis ES (2009) Expres-sion, imprinting, and evolution of rice homologs of the polycomb groupgenes. Mol Plant 2: 711–723

Masiero S, Colombo L, Grini PE, Schnittger A, Kater MM (2011) Theemerging importance of type I MADS box transcription factors for plantreproduction. Plant Cell 23: 865–872

Olsen OA (2004) Nuclear endosperm development in cereals and Arabi-dopsis thaliana. Plant Cell 16(Suppl): S214–S227





Parenicová L, de Folter S, Kieffer M, Horner DS, Favalli C, Busscher J,Cook HE, Ingram RM, Kater MM, Davies B, et al (2003) Molecular andphylogenetic analyses of the complete MADS-box transcription factorfamily in Arabidopsis: New openings to the MADS world. Plant Cell 15:1538–1551

Paul P, Chaturvedi P, Mesihovic A, Ghatak A, Weckwerth W, Schleiff E(2017) Protocol for enrichment of the membrane proteome of maturetomato pollen. Bio Protoc 7: 1–9

Paul P, Chaturvedi P, Selymesi M, Ghatak A, Mesihovic A, Scharf KD,Weckwerth W, Simm S, Schleiff E (2016) The membrane proteome ofmale gametophyte in Solanum lycopersicum. J Proteomics 131: 48–60

Pires ND (2014) Seed evolution: Parental conflicts in a multi-generationalhousehold. Biomol Concepts 5: 71–86

Portereiko MF, Lloyd A, Steffen JG, Punwani JA, Otsuga D, Drews GN(2006) AGL80 is required for central cell and endosperm development inArabidopsis. Plant Cell 18: 1862–1872

Sabelli PA, Larkins BA (2009) The development of endosperm in grasses.Plant Physiol 149: 14–26

Sandhu J, Zhu F, Paul P, Gao T, Dhatt BK, Ge Y, Staswick P, Yu H, WaliaH (2019) PI-Plat: A high-resolution image-based 3D reconstructionmethod to estimate growth dynamics of rice inflorescence traits. bioRxiv835306

Scott RJ, Spielman M, Bailey J, Dickinson HG (1998) Parent-of-origineffects on seed development in Arabidopsis thaliana. Development 125:3329–3341

Singh N, Nakaura Y, Inouchi N, Nishinari K (2007) Fine structure, thermaland viscoelastic properties of starches separated from indica rice culti-vars. Starke 59: 10–20

Smaczniak C, Immink RGH, Angenent GC, Kaufmann K (2012) Devel-opmental and evolutionary diversity of plant MADS-domain factors:Insights from recent studies. Development 139: 3081–3098

Steffen JG, Kang IH, Portereiko MF, Lloyd A, Drews GN (2008) AGL61interacts with AGL80 and is required for central cell development inArabidopsis. Plant Physiol 148: 259–268

Stoute AI, Varenko V, King GJ, Scott RJ, Kurup S (2012) Parental genomeimbalance in Brassica oleracea causes asymmetric triploid block. Plant J71: 503–516

Tanabata T, Shibaya T, Hori K, Ebana K, Yano M (2012) SmartGrain:High-throughput phenotyping software for measuring seed shapethrough image analysis. Plant Physiol 160: 1871–1880

Trapnell C, Pachter L, Salzberg SL (2009) TopHat: Discovering splicejunctions with RNA-Seq. Bioinformatics 25: 1105–1111

Uchiumi T, Okamoto T (2010) Rice fruit development is associated with anincreased IAA content in pollinated ovaries. Planta 232: 579–592

Walia H, Josefsson C, Dilkes B, Kirkbride R, Harada J, Comai L (2009)Dosage-dependent deregulation of an AGAMOUS-LIKE gene clustercontributes to interspecific incompatibility. Curr Biol 19: 1128–1132

Wang L, Yuan J, Ma Y, Jiao W, Ye W, Yang DL, Yi C, Chen ZJ (2018) Riceinterploidy crosses disrupt epigenetic regulation, gene expression, andseed development. Mol Plant 11: 300–314

Yoshihara T, Washida H, Takaiwa F (1996) A 45-bp proximal regioncontaining AACA and GCN4 motif is sufficient to confer endosperm-specific expression of the rice storage protein glutelin gene, GluA-3.FEBS Lett 383: 213–218

Zhang S, Wang D, Zhang H, Skaggs MI, Lloyd A, Ran D, An L,Schumaker KS, Drews GN, Yadegari R (2018) FERTILIZATION-INDEPENDENT SEED-Polycomb Repressive Complex 2 plays a dualrole in regulating type I MADS-box genes in early endosperm devel-opment. Plant Physiol 177: 285–229

Zhao Z, Zhang Y, Liu X, Zhang X, Liu S, Yu X, Ren Y, Zheng X, Zhou K,Jiang L, et al (2013) A role for a dioxygenase in auxin metabolism andreproductive development in rice. Dev Cell 27: 113–122


Paul et al.



MADS78 and MADS79 Are Essential Regulators of Early ...MADS78 and MADS79 Are Essential Regulators of...

Documents

Transcript of MADS78 and MADS79 Are Essential Regulators of Early ...MADS78 and MADS79 Are Essential Regulators of...